In a society with aging population where fewer people of different generations live together, there are increasing risks of people failing to notice deterioration in the health of the elderly living alone or with no one of younger generations in the household, or a degradation in their living functions. Thus, a need exists for a system for efficiently monitoring the condition of residents.
Conventionally, resident monitoring systems are known including devices that monitor the state of utilization of pots, gas, water, electricity and the like; devices that detect passage of someone in front of a sensor installed in the house; and devices that allow a resident to alert people by pushing a button in case of emergency. These devices commonly monitor well-being by issuing notifications to the outside should abnormality develops.
Meanwhile, the elderly may fall and become unable to move, or encounter events requiring emergency care. In these cases, it is often difficult to expect their complete recovery even if treated properly, forcing the person bedridden or in need of nursing care. Thus, in order for the elderly to live an independent life longer, it is desirable to detect signs of deterioration in health or degradation of living functions and to take preventive action, rather than issuing alerts after abnormality has occurred. The conventional monitoring devices, however, do not include such function.
As a monitoring technology for estimating behaviors in everyday life, Patent Literature 1 discloses a subject monitoring system that monitors sounds using a sound sensor device. Patent Literature 1 also discloses a technology that estimates the location of a room in which sound was generated based on an intensity ratio of sounds picked up by a plurality of sound sensors, and that then estimates the cause of the sounds as well as their features.